
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
%load_ext autoreload
%autoreload 2
%matplotlib inline


!pip install -r requirements.txt

Requirement already satisfied: pandas in /Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages (from -r requirements.txt (line 1)) (1.3.5)
Requirement already satisfied: numpy in /Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages (from -r requirements.txt (line 2)) (1.21.5)
Requirement already satisfied: folium in /Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages (from -r requirements.txt (line 3)) (0.12.1.post1)
ERROR: Could not find a version that satisfies the requirement zipfile (from versions: none)
ERROR: No matching distribution found for zipfile


!jupyter nbextension enable --py widgetsnbextension
!jupyter trust Airbnb_ML.ipynb

Enabling notebook extension jupyter-js-widgets/extension...
      - Validating: OK
Signing notebook: Airbnb_ML.ipynb


import os

import sys 
print("Python version: {}". format(sys.version))

import IPython
from IPython import display
from IPython.display import Markdown, display, HTML
print("IPython version: {}". format(IPython.__version__))

import numpy as np
print("NumPy version: {}". format(np.__version__))

import pandas as pd
print("pandas version: {}". format(pd.__version__))

import matplotlib
import matplotlib.pyplot as plt
print("matplotlib version: {}". format(matplotlib.__version__))

import seaborn as sns
print("seaborn version : {}". format(sns.__version__))
sns.set()

import missingno as msno
print("missingno version : {}". format(msno.__version__))

import pandas_profiling
print("pandas_profiling version : {}". format(pandas_profiling.__version__))

import folium
print("folium version : {}". format(folium.__version__))

import wordcloud
print("wordcloud version : {}". format(wordcloud.__version__))

import ipywidgets
print("ipywidgets version : {}". format(ipywidgets.__version__))

import sklearn
print("scikit-learn version : {}". format(sklearn.__version__))

print("====================================")

Python version: 3.9.2 (v3.9.2:1a79785e3e, Feb 19 2021, 09:06:10) 
[Clang 6.0 (clang-600.0.57)]
IPython version: 7.20.0
NumPy version: 1.21.5
pandas version: 1.3.5
matplotlib version: 3.5.1
seaborn version : 0.11.2
missingno version : 0.5.0
pandas_profiling version : 3.1.0
folium version : 0.12.1.post1
wordcloud version : 1.8.1
ipywidgets version : 7.6.5
scikit-learn version : 1.0.2
====================================


import zipfile
 
zip_file = "./dataset/airbnb_compressed.zip"

try:
    with zipfile.ZipFile(zip_file) as z:
        z.extractall("./dataset")
        print("Extracted all")
except:
    print("Invalid file")

Extracted all


# df_main => cleaned summary of the dataset, good for visualisations
df_main = pd.read_csv('./dataset/airbnb_paris.csv', index_col='id')

# df_listings => detailed dataset with all collected columns
df_listings = pd.read_csv('./dataset/listings.csv', index_col='id')


data_folder = "dataset"
file_name = "Inside_Airbnb_Data_Dictionary.csv"
dictionary_file_path = os.path.join(".", data_folder, file_name)
separateur = ";"
pd.options.display.max_rows = 80
df_dictionary = pd.read_csv(dictionary_file_path, sep=separateur, index_col="Field")
df_dictionary[["Type", "Description"]]


df_main.head()


df_main.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 49429 entries, 5396 to 53712667
Data columns (total 17 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   name                            49384 non-null  object 
 1   host_id                         49429 non-null  int64  
 2   host_name                       49397 non-null  object 
 3   neighbourhood_group             0 non-null      float64
 4   neighbourhood                   49429 non-null  object 
 5   latitude                        49429 non-null  float64
 6   longitude                       49429 non-null  float64
 7   room_type                       49429 non-null  object 
 8   price                           49429 non-null  int64  
 9   minimum_nights                  49429 non-null  int64  
 10  number_of_reviews               49429 non-null  int64  
 11  last_review                     38767 non-null  object 
 12  reviews_per_month               38767 non-null  float64
 13  calculated_host_listings_count  49429 non-null  int64  
 14  availability_365                49429 non-null  int64  
 15  number_of_reviews_ltm           49429 non-null  int64  
 16  license                         28122 non-null  object 
dtypes: float64(4), int64(7), object(6)
memory usage: 6.8+ MB


df_main.describe()


# https://python-visualization.github.io/folium/
import folium, folium.plugins 
paris_location = [48.86, 2.34]
map = folium.Map(location=paris_location, zoom_start=12)
folium.plugins.FastMarkerCluster(data=df_main[['latitude', 'longitude']]).add_to(map)
map.save( 'map.html')
map


def generate_wordcloud(df,col):
    content = ''
    stopwords = set(STOPWORDS)
    

    for val in df[col].values:
        
        val = str(val)
        tokens = val.split()
        for i in range(len(tokens)):
            tokens[i] = tokens[i].lower()
        
        content += " ".join(tokens)+" "

    wordcloud = WordCloud(width = 800, height = 800,
                    background_color ='white',
                    stopwords = stopwords,
                    min_font_size = 10).generate(content)
    
    # plot the WordCloud image
    
    plt.figure(figsize = (8, 8), facecolor = None)
    plt.imshow(wordcloud)
    plt.axis("off")
    plt.tight_layout(pad = 0)


%matplotlib inline


from wordcloud import WordCloud, STOPWORDS

generate_wordcloud(df_main,'name')

# the bigger the word, the more frequent it is in this field


generate_wordcloud(df_main,'neighbourhood')


df_listings.head()


df_listings.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 49429 entries, 5396 to 53712667
Data columns (total 73 columns):
 #   Column                                        Non-Null Count  Dtype  
---  ------                                        --------------  -----  
 0   listing_url                                   49429 non-null  object 
 1   scrape_id                                     49429 non-null  int64  
 2   last_scraped                                  49429 non-null  object 
 3   name                                          49384 non-null  object 
 4   description                                   48706 non-null  object 
 5   neighborhood_overview                         29455 non-null  object 
 6   picture_url                                   49428 non-null  object 
 7   host_id                                       49429 non-null  int64  
 8   host_url                                      49429 non-null  object 
 9   host_name                                     49397 non-null  object 
 10  host_since                                    49397 non-null  object 
 11  host_location                                 49303 non-null  object 
 12  host_about                                    25075 non-null  object 
 13  host_response_time                            21820 non-null  object 
 14  host_response_rate                            21820 non-null  object 
 15  host_acceptance_rate                          24605 non-null  object 
 16  host_is_superhost                             49397 non-null  object 
 17  host_thumbnail_url                            49397 non-null  object 
 18  host_picture_url                              49397 non-null  object 
 19  host_neighbourhood                            35037 non-null  object 
 20  host_listings_count                           49397 non-null  float64
 21  host_total_listings_count                     49397 non-null  float64
 22  host_verifications                            49429 non-null  object 
 23  host_has_profile_pic                          49397 non-null  object 
 24  host_identity_verified                        49397 non-null  object 
 25  neighbourhood                                 29455 non-null  object 
 26  neighbourhood_cleansed                        49429 non-null  object 
 27  neighbourhood_group_cleansed                  0 non-null      float64
 28  latitude                                      49429 non-null  float64
 29  longitude                                     49429 non-null  float64
 30  property_type                                 49429 non-null  object 
 31  room_type                                     49429 non-null  object 
 32  accommodates                                  49429 non-null  int64  
 33  bathrooms                                     0 non-null      float64
 34  bathrooms_text                                49274 non-null  object 
 35  bedrooms                                      40293 non-null  float64
 36  beds                                          47485 non-null  float64
 37  amenities                                     49429 non-null  object 
 38  price                                         49429 non-null  object 
 39  minimum_nights                                49429 non-null  int64  
 40  maximum_nights                                49429 non-null  int64  
 41  minimum_minimum_nights                        49424 non-null  float64
 42  maximum_minimum_nights                        49424 non-null  float64
 43  minimum_maximum_nights                        49424 non-null  float64
 44  maximum_maximum_nights                        49424 non-null  float64
 45  minimum_nights_avg_ntm                        49424 non-null  float64
 46  maximum_nights_avg_ntm                        49424 non-null  float64
 47  calendar_updated                              0 non-null      float64
 48  has_availability                              49429 non-null  object 
 49  availability_30                               49429 non-null  int64  
 50  availability_60                               49429 non-null  int64  
 51  availability_90                               49429 non-null  int64  
 52  availability_365                              49429 non-null  int64  
 53  calendar_last_scraped                         49429 non-null  object 
 54  number_of_reviews                             49429 non-null  int64  
 55  number_of_reviews_ltm                         49429 non-null  int64  
 56  number_of_reviews_l30d                        49429 non-null  int64  
 57  first_review                                  38767 non-null  object 
 58  last_review                                   38767 non-null  object 
 59  review_scores_rating                          38767 non-null  float64
 60  review_scores_accuracy                        38119 non-null  float64
 61  review_scores_cleanliness                     38124 non-null  float64
 62  review_scores_checkin                         38105 non-null  float64
 63  review_scores_communication                   38117 non-null  float64
 64  review_scores_location                        38103 non-null  float64
 65  review_scores_value                           38100 non-null  float64
 66  license                                       28122 non-null  object 
 67  instant_bookable                              49429 non-null  object 
 68  calculated_host_listings_count                49429 non-null  int64  
 69  calculated_host_listings_count_entire_homes   49429 non-null  int64  
 70  calculated_host_listings_count_private_rooms  49429 non-null  int64  
 71  calculated_host_listings_count_shared_rooms   49429 non-null  int64  
 72  reviews_per_month                             38767 non-null  float64
dtypes: float64(23), int64(16), object(34)
memory usage: 27.9+ MB


df_listings.describe()


df_main.columns

Index(['name', 'host_id', 'host_name', 'neighbourhood_group', 'neighbourhood',
       'latitude', 'longitude', 'room_type', 'price', 'minimum_nights',
       'number_of_reviews', 'last_review', 'reviews_per_month',
       'calculated_host_listings_count', 'availability_365',
       'number_of_reviews_ltm', 'license'],
      dtype='object')


# display columns that are not already in df_main
for elt in df_listings.columns:
    if elt not in df_main.columns:
        print(elt, "(",df_listings[elt].dtype, ")", end=', ')

listing_url ( object ), scrape_id ( int64 ), last_scraped ( object ), description ( object ), neighborhood_overview ( object ), picture_url ( object ), host_url ( object ), host_since ( object ), host_location ( object ), host_about ( object ), host_response_time ( object ), host_response_rate ( object ), host_acceptance_rate ( object ), host_is_superhost ( object ), host_thumbnail_url ( object ), host_picture_url ( object ), host_neighbourhood ( object ), host_listings_count ( float64 ), host_total_listings_count ( float64 ), host_verifications ( object ), host_has_profile_pic ( object ), host_identity_verified ( object ), neighbourhood_cleansed ( object ), neighbourhood_group_cleansed ( float64 ), property_type ( object ), accommodates ( int64 ), bathrooms ( float64 ), bathrooms_text ( object ), bedrooms ( float64 ), beds ( float64 ), amenities ( object ), maximum_nights ( int64 ), minimum_minimum_nights ( float64 ), maximum_minimum_nights ( float64 ), minimum_maximum_nights ( float64 ), maximum_maximum_nights ( float64 ), minimum_nights_avg_ntm ( float64 ), maximum_nights_avg_ntm ( float64 ), calendar_updated ( float64 ), has_availability ( object ), availability_30 ( int64 ), availability_60 ( int64 ), availability_90 ( int64 ), calendar_last_scraped ( object ), number_of_reviews_l30d ( int64 ), first_review ( object ), review_scores_rating ( float64 ), review_scores_accuracy ( float64 ), review_scores_cleanliness ( float64 ), review_scores_checkin ( float64 ), review_scores_communication ( float64 ), review_scores_location ( float64 ), review_scores_value ( float64 ), instant_bookable ( object ), calculated_host_listings_count_entire_homes ( int64 ), calculated_host_listings_count_private_rooms ( int64 ), calculated_host_listings_count_shared_rooms ( int64 ),


# Data dictionary : https://docs.google.com/spreadsheets/d/1iWCNJcSutYqpULSQHlNyGInUvHg2BoUGoNRIGa6Szc4/edit#gid=982310896
listings_cols_to_keep = ["host_is_superhost",
                        "accommodates", 
                        "bedrooms",
                        "beds",
                        "availability_60",
                        "number_of_reviews_l30d",
                        "availability_90",
                        "review_scores_accuracy",
                        "review_scores_cleanliness",
                        "review_scores_checkin",
                        "review_scores_communication",
                        "review_scores_location",
                        "review_scores_value"
                        # TODO Find More to add
                        ]
df = pd.merge(df_main, df_listings[listings_cols_to_keep], on='id', how='left')

df


df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 49429 entries, 5396 to 53712667
Data columns (total 30 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   name                            49384 non-null  object 
 1   host_id                         49429 non-null  int64  
 2   host_name                       49397 non-null  object 
 3   neighbourhood_group             0 non-null      float64
 4   neighbourhood                   49429 non-null  object 
 5   latitude                        49429 non-null  float64
 6   longitude                       49429 non-null  float64
 7   room_type                       49429 non-null  object 
 8   price                           49429 non-null  int64  
 9   minimum_nights                  49429 non-null  int64  
 10  number_of_reviews               49429 non-null  int64  
 11  last_review                     38767 non-null  object 
 12  reviews_per_month               38767 non-null  float64
 13  calculated_host_listings_count  49429 non-null  int64  
 14  availability_365                49429 non-null  int64  
 15  number_of_reviews_ltm           49429 non-null  int64  
 16  license                         28122 non-null  object 
 17  host_is_superhost               49397 non-null  object 
 18  accommodates                    49429 non-null  int64  
 19  bedrooms                        40293 non-null  float64
 20  beds                            47485 non-null  float64
 21  availability_60                 49429 non-null  int64  
 22  number_of_reviews_l30d          49429 non-null  int64  
 23  availability_90                 49429 non-null  int64  
 24  review_scores_accuracy          38119 non-null  float64
 25  review_scores_cleanliness       38124 non-null  float64
 26  review_scores_checkin           38105 non-null  float64
 27  review_scores_communication     38117 non-null  float64
 28  review_scores_location          38103 non-null  float64
 29  review_scores_value             38100 non-null  float64
dtypes: float64(12), int64(11), object(7)
memory usage: 11.7+ MB


df.describe()


msno.matrix(df)

<AxesSubplot:>


df['neighbourhood_group'].unique()

array([nan])


del df['neighbourhood_group']


df['license'].isna().sum() / len(df)

0.431062736450262


del df['license']


print("The last_review column (The date of the last/newest review) has {:.2%} missing data." \
      .format(df['last_review'].isna().sum() / len(df)))

The last_review column (The date of the last/newest review) has 21.57% missing data.


del df['last_review']


df['host_name'] =  df['host_name'].fillna("NO_NAME")


def process_empty(x):
    x = x.replace(' ','')
    return x if x!='' else np.NaN

for col in df.columns :
    try:
        df[col] = df[col].apply(process_empty)
    except:
        pass


data_type = {
    'int':['id','price','host_id','minimum_nights','calculated_host_listings_count',
           'availability_365','number_of_reviews_ltm'],
    'float':['latitude','longitude','reviews_per_month'],
    'datetime':['last_review'],
    'object':['name','host_name','room_type']
}


for col in df.columns:
    for key, values in data_type.items():

        if col in values:
            if key == 'datetime':
                df[col] = pd.to_datetime(df[col])
                df['last_review'] = df['last_review'].dt.strftime('%d-%m-%Y')                
            else:
                df[col] = df[col].astype(key)


df['reviews_per_month'] = df['reviews_per_month'].fillna(0)


df['host_is_superhost'] = df['host_is_superhost'].fillna('f')


# df_listings[df_listings['bedrooms'].isna()]['listing_url']
# quand bedrooms = NaN, pas de chambre, ex studio 
df['bedrooms'] = df['bedrooms'].fillna(0)
df['beds']     = df['beds'].fillna(1)


df['name'] = df['name'].fillna(" ")


scores = ['accuracy', 'cleanliness', 'checkin', 'communication', 'location', 'value']
df['review_scores_' + scores[0]].isna().sum() / df.shape[0]

for elt in scores: 
    index_with_nan = df.index[df[['review_scores_' + elt]].isnull().any(axis=1)]
    df.drop(index_with_nan,axis=0, inplace=True)


msno.matrix(df)

<AxesSubplot:>


df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 38095 entries, 5396 to 53669791
Data columns (total 27 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   name                            38095 non-null  object 
 1   host_id                         38095 non-null  int64  
 2   host_name                       38095 non-null  object 
 3   neighbourhood                   38095 non-null  object 
 4   latitude                        38095 non-null  float64
 5   longitude                       38095 non-null  float64
 6   room_type                       38095 non-null  object 
 7   price                           38095 non-null  int64  
 8   minimum_nights                  38095 non-null  int64  
 9   number_of_reviews               38095 non-null  int64  
 10  reviews_per_month               38095 non-null  float64
 11  calculated_host_listings_count  38095 non-null  int64  
 12  availability_365                38095 non-null  int64  
 13  number_of_reviews_ltm           38095 non-null  int64  
 14  host_is_superhost               38095 non-null  object 
 15  accommodates                    38095 non-null  int64  
 16  bedrooms                        38095 non-null  float64
 17  beds                            38095 non-null  float64
 18  availability_60                 38095 non-null  int64  
 19  number_of_reviews_l30d          38095 non-null  int64  
 20  availability_90                 38095 non-null  int64  
 21  review_scores_accuracy          38095 non-null  float64
 22  review_scores_cleanliness       38095 non-null  float64
 23  review_scores_checkin           38095 non-null  float64
 24  review_scores_communication     38095 non-null  float64
 25  review_scores_location          38095 non-null  float64
 26  review_scores_value             38095 non-null  float64
dtypes: float64(11), int64(11), object(5)
memory usage: 8.1+ MB


df.describe()


f, ax = plt.subplots(figsize=(13, 7))

sns.boxplot(data=df, x='room_type', y='price', ax=ax)
ax.set_ylabel('price (€)')
ax.set_xlabel('room type')
ax.set_xticklabels(ax.get_xticklabels(), rotation=90)

_ = f.suptitle('price by room_type', fontsize=20)


target = df['price']
outlier_threshold = 500
condition = target > outlier_threshold
nb_outliers = target[condition].count()
print("{} appartments over {} euros out of a total of {}".format(nb_outliers, outlier_threshold, target.shape[0]))
print("i.e. {:.2%}".format(nb_outliers/target.shape[0]))

500 appartments over 500 euros out of a total of 38095
i.e. 1.31%


condition = (target > 0) & (target < outlier_threshold)
df = df[condition]


f, ax = plt.subplots(figsize=(13, 7))

sns.boxplot(data=df, x='room_type', y='price', ax=ax)
ax.set_ylabel('price (€)')
ax.set_xlabel('room type')
ax.set_xticklabels(ax.get_xticklabels(), rotation=90)

_ = f.suptitle('price by room_type', fontsize=20)


f, ax = plt.subplots(figsize=(13, 7))

sns.violinplot(data=df, x='room_type', y='price', ax=ax)
ax.set_ylabel('price (€)')
ax.set_xlabel('room type')
ax.set_xticklabels(ax.get_xticklabels(), rotation=90)

_ = f.suptitle('price by room_type', fontsize=20)


from pandas_profiling import ProfileReport
# Caution: Profiling the df dataset is time consuming
profile = ProfileReport(df, title='df Profiling Report', html={'style':{'full_width':True}}, \
                        vars = {'cat': {'check_composition' : False}})
profile.to_notebook_iframe()

Summarize dataset: 100%|██████████| 570/570 [02:11<00:00,  4.34it/s, Completed]                                                             
Generate report structure: 100%|██████████| 1/1 [00:15<00:00, 15.15s/it]
Render HTML: 100%|██████████| 1/1 [00:23<00:00, 23.56s/it]


profile.to_file("airbnb_data_exp.html")

Export report to file: 100%|██████████| 1/1 [00:00<00:00,  2.53it/s]


f, ax = plt.subplots(figsize=(13, 7))
sns.histplot(data=df, x="price", hue='room_type', ax=ax)

<AxesSubplot:xlabel='price', ylabel='Count'>


sns.displot(data=df, x='price', col='room_type', kind="kde", height=6, aspect=0.5)

<seaborn.axisgrid.FacetGrid at 0x7fb753bb32b0>


df_Nb_rooms_by_type = df['room_type'].value_counts()
df_Nb_rooms_by_type

Entirehome/apt    31643
Privateroom        4954
Hotelroom           709
Sharedroom          215
Name: room_type, dtype: int64


colors = sns.color_palette('pastel')
explode = [0.02, 0.02, 0.3, 0.3]
df_Nb_rooms_by_type.plot.pie(colors = colors,
                             fontsize = 14,
                             autopct = '%0.0f%%',
                             explode = explode,                             
                             figsize=(8, 8))

<AxesSubplot:ylabel='room_type'>


plt.figure(figsize=(13,7))
sns.scatterplot(data=df, x='price', y='neighbourhood', hue='room_type')

<AxesSubplot:xlabel='price', ylabel='neighbourhood'>


# We use the result of the profiling to select the columns

Colonnes_retenues = ['price',
                     'accommodates',
                     'bedrooms',
                     'beds',
                     'availability_365',
                     'availability_90',
                     'availability_60',
                     'calculated_host_listings_count',
                     'minimum_nights',
                     'longitude',
                     'reviews_per_month', 
                     'number_of_reviews_l30d',
                     'number_of_reviews_ltm',
                     'number_of_reviews',
                     'latitude']

fig, ax = plt.subplots(figsize=(15,15)) 
sns.heatmap(df[Colonnes_retenues].corr()**2, 
            annot=True, square=True, cmap='coolwarm', annot_kws={'size': 14}, ax=ax)
plt.title("Matrix of the determination coefficients of the dataframe df")
plt.show()


Colonnes_retenues = ['price',
                     'accommodates',
                     'bedrooms',
                     'beds',
                     'availability_365',
                     'availability_90',
                     'availability_60']

fig, ax = plt.subplots(figsize=(10,10)) 
sns.heatmap(df[Colonnes_retenues].corr()**2, 
            annot=True, square=True, cmap='coolwarm', annot_kws={'size': 14}, ax=ax)
plt.title("Matrix of the determination coefficients of the dataframe df")
plt.show()


df.columns

Index(['name', 'host_id', 'host_name', 'neighbourhood', 'latitude',
       'longitude', 'room_type', 'price', 'minimum_nights',
       'number_of_reviews', 'reviews_per_month',
       'calculated_host_listings_count', 'availability_365',
       'number_of_reviews_ltm', 'host_is_superhost', 'accommodates',
       'bedrooms', 'beds', 'availability_60', 'number_of_reviews_l30d',
       'availability_90', 'review_scores_accuracy',
       'review_scores_cleanliness', 'review_scores_checkin',
       'review_scores_communication', 'review_scores_location',
       'review_scores_value'],
      dtype='object')


df_model = df.drop(columns=['name', 'host_id', 'host_name'])


X = df_model.drop(['price'], axis=1)
y = df_model['price']


X.columns[X.dtypes == 'object']

Index(['neighbourhood', 'room_type', 'host_is_superhost'], dtype='object')


from sklearn.preprocessing import OrdinalEncoder
for object_col in X.columns[X.dtypes == 'object']:
    enc = OrdinalEncoder()
    enc.fit(X[[object_col]])
    enc.categories_
    X[object_col] = enc.transform(X[[object_col]])


from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42)


X_train.describe()


X_train.isna().sum()

neighbourhood                     0
latitude                          0
longitude                         0
room_type                         0
minimum_nights                    0
number_of_reviews                 0
reviews_per_month                 0
calculated_host_listings_count    0
availability_365                  0
number_of_reviews_ltm             0
host_is_superhost                 0
accommodates                      0
bedrooms                          0
beds                              0
availability_60                   0
number_of_reviews_l30d            0
availability_90                   0
review_scores_accuracy            0
review_scores_cleanliness         0
review_scores_checkin             0
review_scores_communication       0
review_scores_location            0
review_scores_value               0
dtype: int64


from sklearn.linear_model import LinearRegression
from sklearn.linear_model import Ridge
from sklearn.linear_model import Lasso
from sklearn.linear_model import LassoCV
from sklearn.linear_model import ElasticNetCV
from sklearn.svm import SVR
from sklearn.ensemble import GradientBoostingRegressor
from sklearn.ensemble import RandomForestRegressor
from xgboost import XGBRegressor
from sklearn.ensemble import HistGradientBoostingRegressor
from lightgbm import LGBMRegressor

Liste_modeles = {
                    "LinearRegression": LinearRegression(),
                    "Ridge": Ridge(random_state=0),
                    "Lasso": Lasso(random_state=0),
                    "LassoCV": LassoCV(cv=5, random_state=0),
                    "ElasticNetCV": ElasticNetCV(cv=5, random_state=0),
                    "SVR": SVR(),
                    "RandomForest": RandomForestRegressor(random_state=0),
                    "GradientBoosting": GradientBoostingRegressor(random_state=0),
                    "xgboost": XGBRegressor(random_state=0),
                    "HistGradientBoosting": HistGradientBoostingRegressor(random_state=0),
                    "LGBM": LGBMRegressor()
                }


from sklearn.metrics import mean_absolute_percentage_error, mean_absolute_error, r2_score

Eval_model = {}
Eval_model["model"] = []
Eval_model["R2"] = []
Eval_model["MAE"] = []
Eval_model["MAPE"] = []

for model in Liste_modeles:
    print("---------- {} ----------".format(model))
    reg = Liste_modeles[model].fit(X_train, y_train)
    y_pred = reg.predict(X_test)
    Eval_model["model"].append(model)
    Eval_model["R2"].append(r2_score(y_test, y_pred))
    Eval_model["MAE"].append(mean_absolute_error(y_test, y_pred))
    Eval_model["MAPE"].append(mean_absolute_percentage_error(y_test, y_pred))
    
print("")
df_eval_model = pd.DataFrame(Eval_model).set_index("model")
df_eval_model.sort_values(by="R2", ascending=False, inplace=True)
df_eval_model

---------- LinearRegression ----------
---------- Ridge ----------
---------- Lasso ----------
---------- LassoCV ----------
---------- ElasticNetCV ----------
---------- SVR ----------
---------- RandomForest ----------
---------- GradientBoosting ----------
---------- xgboost ----------
---------- HistGradientBoosting ----------
---------- LGBM ----------


from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold

n_splits = 3
cv = KFold(n_splits=n_splits)
The_best_models = df_eval_model.head(5).index.values.tolist()

cross_valid = {}
for model in The_best_models:
    reg = Liste_modeles[model].fit(X_train, y_train)
    results = cross_val_score(reg, X, y, cv=cv)
    cross_valid[model] = results

df_cross_valid = pd.DataFrame(cross_valid)
moyenne = df_cross_valid.mean()
df_cross_valid = df_cross_valid.T
df_cross_valid["mean"] = moyenne
df_cross_valid.sort_values(by="mean", ascending=False, inplace=True)
df_cross_valid


The_best_model = df_cross_valid.head(1).index.values.tolist()[0]
print("The best model is {}.".format(The_best_model))

The best model is LGBM.


!
!jupyter nbconvert --to HTML Airbnb_ML.ipynb

^C
Fatal Python error: init_import_site: Failed to import the site module
Python runtime state: initialized
Traceback (most recent call last):
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site.py", line 589, in <module>
    main()
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site.py", line 576, in main
    known_paths = addsitepackages(known_paths)
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site.py", line 359, in addsitepackages
    addsitedir(sitedir, known_paths)
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site.py", line 208, in addsitedir
    addpackage(sitedir, name, known_paths)
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site.py", line 169, in addpackage
    exec(line)
  File "<string>", line 1, in <module>
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/importlib/util.py", line 2, in <module>
    from . import abc
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/importlib/abc.py", line 17, in <module>
    from typing import Protocol, runtime_checkable
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/typing.py", line 26, in <module>
    import re as stdlib_re  # Avoid confusion with the re we export.
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/re.py", line 125, in <module>
    import sre_compile
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/sre_compile.py", line 14, in <module>
    import sre_parse
  File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/sre_parse.py", line 15, in <module>
    from sre_constants import *
  File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
  File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 786, in exec_module
  File "<frozen importlib._bootstrap_external>", line 881, in get_code
  File "<frozen importlib._bootstrap_external>", line 980, in get_data
KeyboardInterrupt

	name	host_id	host_name	neighbourhood_group	neighbourhood	latitude	longitude	room_type	price	minimum_nights	number_of_reviews	last_review	reviews_per_month	calculated_host_listings_count	availability_365	number_of_reviews_ltm	license
id
5396	Explore the heart of old Paris	7903	Borzou	NaN	Hôtel-de-Ville	48.85247	2.35835	Entire home/apt	102	2	273	2021-12-08	1.80	1	58	42	7510402838018
7397	MARAIS - 2ROOMS APT - 2/4 PEOPLE	2626	Franck	NaN	Hôtel-de-Ville	48.85909	2.35315	Entire home/apt	112	10	288	2021-12-05	2.22	2	209	19	7510400829623
7964	Large & sunny flat with balcony !	22155	Anaïs	NaN	Opéra	48.87417	2.34245	Entire home/apt	130	6	6	2015-09-14	0.04	1	344	0	7510903576564
9359	Cozy, Central Paris: WALK or VELIB EVERYWHERE !	28422	Bernadette	NaN	Louvre	48.86006	2.34863	Entire home/apt	75	180	0	NaN	NaN	1	117	0	Available with a mobility lease only ("bail mo...
9952	Paris petit coin douillet	33534	Elisabeth	NaN	Popincourt	48.86373	2.37093	Entire home/apt	81	4	33	2021-10-22	0.31	1	260	7	7511101582862

	host_id	neighbourhood_group	latitude	longitude	price	minimum_nights	number_of_reviews	reviews_per_month	calculated_host_listings_count	availability_365	number_of_reviews_ltm
count	4.942900e+04	0.0	49429.000000	49429.000000	49429.000000	49429.000000	49429.000000	38767.000000	49429.000000	49429.000000	49429.000000
mean	9.896725e+07	NaN	48.864008	2.344914	129.891986	109.405086	21.636590	0.796471	10.652188	99.513059	3.635801
std	1.208700e+08	NaN	0.018181	0.033072	217.186069	168.774442	46.339395	1.191827	35.528471	134.092923	13.033774
min	2.626000e+03	NaN	48.812580	2.225720	0.000000	1.000000	0.000000	0.010000	1.000000	0.000000	0.000000
25%	1.340742e+07	NaN	48.850820	2.324110	61.000000	2.000000	1.000000	0.120000	1.000000	0.000000	0.000000
50%	3.883365e+07	NaN	48.865320	2.347900	90.000000	4.000000	6.000000	0.370000	1.000000	3.000000	0.000000
75%	1.470187e+08	NaN	48.878530	2.369270	140.000000	365.000000	21.000000	1.000000	2.000000	198.000000	3.000000
max	4.350258e+08	NaN	48.904860	2.472030	10250.000000	9999.000000	1809.000000	50.860000	252.000000	365.000000	1705.000000

	listing_url	scrape_id	last_scraped	name	description	neighborhood_overview	picture_url	host_id	host_url	host_name	...	review_scores_communication	review_scores_location	review_scores_value	license	instant_bookable	calculated_host_listings_count	calculated_host_listings_count_entire_homes	calculated_host_listings_count_private_rooms	calculated_host_listings_count_shared_rooms	reviews_per_month
id
5396	https://www.airbnb.com/rooms/5396	20211207182322	2021-12-09	Explore the heart of old Paris	Cozy, well-appointed and graciously designed s...	You are within walking distance to the Louvre,...	https://a0.muscache.com/pictures/52413/f9bf76f...	7903	https://www.airbnb.com/users/show/7903	Borzou	...	4.81	4.96	4.53	7510402838018	f	1	1	0	0	1.80
7397	https://www.airbnb.com/rooms/7397	20211207182322	2021-12-09	MARAIS - 2ROOMS APT - 2/4 PEOPLE	VERY CONVENIENT, WITH THE BEST LOCATION !<br /...	NaN	https://a0.muscache.com/pictures/67928287/330b...	2626	https://www.airbnb.com/users/show/2626	Franck	...	4.88	4.92	4.71	7510400829623	f	2	2	0	0	2.22
7964	https://www.airbnb.com/rooms/7964	20211207182322	2021-12-08	Large & sunny flat with balcony !	Very large & nice apartment all for you! <br /...	NaN	https://a0.muscache.com/pictures/4471349/6fb3d...	22155	https://www.airbnb.com/users/show/22155	Anaïs	...	5.00	5.00	5.00	7510903576564	f	1	1	0	0	0.04
9359	https://www.airbnb.com/rooms/9359	20211207182322	2021-12-09	Cozy, Central Paris: WALK or VELIB EVERYWHERE !	Location! Location! Location! Just bring your ...	NaN	https://a0.muscache.com/pictures/c2965945-061f...	28422	https://www.airbnb.com/users/show/28422	Bernadette	...	NaN	NaN	NaN	Available with a mobility lease only ("bail mo...	f	1	1	0	0	NaN
9952	https://www.airbnb.com/rooms/9952	20211207182322	2021-12-09	Paris petit coin douillet	Je suis une dame retraitée, qui propose un agr...	Vibrant neighborhood, full of bars, cafés, fre...	https://a0.muscache.com/pictures/ae822d16-74d2...	33534	https://www.airbnb.com/users/show/33534	Elisabeth	...	4.91	4.91	4.94	7511101582862	f	1	1	0	0	0.31

	scrape_id	host_id	host_listings_count	host_total_listings_count	neighbourhood_group_cleansed	latitude	longitude	accommodates	bathrooms	bedrooms	...	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value	calculated_host_listings_count	calculated_host_listings_count_entire_homes	calculated_host_listings_count_private_rooms	calculated_host_listings_count_shared_rooms	reviews_per_month
count	4.942900e+04	4.942900e+04	49397.000000	49397.000000	0.0	49429.000000	49429.000000	49429.000000	0.0	40293.000000	...	38124.00000	38105.000000	38117.000000	38103.000000	38100.000000	49429.000000	49429.000000	49429.000000	49429.000000	38767.000000
mean	2.021121e+13	9.896725e+07	13.513149	13.513149	NaN	48.864008	2.344914	3.044346	NaN	1.369990	...	4.59638	4.802476	4.813442	4.805913	4.618081	10.652188	9.945579	0.460904	0.019604	0.796471
std	0.000000e+00	1.208700e+08	83.562486	83.562486	NaN	0.018181	0.033072	1.622670	NaN	1.015249	...	0.53207	0.399383	0.401767	0.348673	0.464296	35.528471	35.404664	2.101099	0.306851	1.191827
min	2.021121e+13	2.626000e+03	0.000000	0.000000	NaN	48.812580	2.225720	0.000000	NaN	1.000000	...	0.00000	0.000000	0.000000	0.000000	0.000000	1.000000	0.000000	0.000000	0.000000	0.010000
25%	2.021121e+13	1.340742e+07	1.000000	1.000000	NaN	48.850820	2.324110	2.000000	NaN	1.000000	...	4.46000	4.760000	4.790000	4.750000	4.500000	1.000000	1.000000	0.000000	0.000000	0.120000
50%	2.021121e+13	3.883365e+07	1.000000	1.000000	NaN	48.865320	2.347900	2.000000	NaN	1.000000	...	4.75000	4.930000	4.950000	4.920000	4.720000	1.000000	1.000000	0.000000	0.000000	0.370000
75%	2.021121e+13	1.470187e+08	2.000000	2.000000	NaN	48.878530	2.369270	4.000000	NaN	2.000000	...	4.97000	5.000000	5.000000	5.000000	4.900000	2.000000	1.000000	0.000000	0.000000	1.000000
max	2.021121e+13	4.350258e+08	1288.000000	1288.000000	NaN	48.904860	2.472030	16.000000	NaN	50.000000	...	5.00000	5.000000	5.000000	5.000000	5.000000	252.000000	249.000000	30.000000	10.000000	50.860000

	host_id	neighbourhood_group	latitude	longitude	price	minimum_nights	number_of_reviews	reviews_per_month	calculated_host_listings_count	availability_365	...	beds	availability_60	number_of_reviews_l30d	availability_90	review_scores_accuracy	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value
count	4.942900e+04	0.0	49429.000000	49429.000000	49429.000000	49429.000000	49429.000000	38767.000000	49429.000000	49429.000000	...	47485.000000	49429.000000	49429.000000	49429.000000	38119.000000	38124.00000	38105.000000	38117.000000	38103.000000	38100.000000
mean	9.896725e+07	NaN	48.864008	2.344914	129.891986	109.405086	21.636590	0.796471	10.652188	99.513059	...	1.729367	13.859698	0.458051	23.360922	4.760520	4.59638	4.802476	4.813442	4.805913	4.618081
std	1.208700e+08	NaN	0.018181	0.033072	217.186069	168.774442	46.339395	1.191827	35.528471	134.092923	...	1.354777	20.322719	1.391720	32.375636	0.422341	0.53207	0.399383	0.401767	0.348673	0.464296
min	2.626000e+03	NaN	48.812580	2.225720	0.000000	1.000000	0.000000	0.010000	1.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	0.000000	0.00000	0.000000	0.000000	0.000000	0.000000
25%	1.340742e+07	NaN	48.850820	2.324110	61.000000	2.000000	1.000000	0.120000	1.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	4.700000	4.46000	4.760000	4.790000	4.750000	4.500000
50%	3.883365e+07	NaN	48.865320	2.347900	90.000000	4.000000	6.000000	0.370000	1.000000	3.000000	...	1.000000	0.000000	0.000000	0.000000	4.880000	4.75000	4.930000	4.950000	4.920000	4.720000
75%	1.470187e+08	NaN	48.878530	2.369270	140.000000	365.000000	21.000000	1.000000	2.000000	198.000000	...	2.000000	27.000000	0.000000	51.000000	5.000000	4.97000	5.000000	5.000000	5.000000	4.900000
max	4.350258e+08	NaN	48.904860	2.472030	10250.000000	9999.000000	1809.000000	50.860000	252.000000	365.000000	...	90.000000	60.000000	62.000000	90.000000	5.000000	5.00000	5.000000	5.000000	5.000000	5.000000

Airbnb Paris dataset analysis¶

Datacamp 2021¶

Libraries¶

Load Data¶

Exploratory data analysis¶

Data dictionary¶

The data from df_main¶

Location of the apartments for rent¶

Textual variables¶

column "name"¶

Column "neighbourhood"¶

The data from df_listings¶

Construction of the dataset that will be used for the analysis¶

Cleaning the data set¶

Dealing with Missing Values¶

Columns deletion¶

Neighbourhood_group¶

License¶

last_review¶

Replacement of missing data with default values¶

Outliers processing¶

The variable 'price' (target variable)¶

Univariate Analysis¶

Profiling the dataset df¶

Target variable description¶

Correlation of 'price' with other variables¶

Feature Engineering¶

Preprocessing¶

Train and test data¶

Evaluation of regression models¶

Selection of best models¶

Cross validation¶

Model optimisation¶

Explainability¶

Conclusion¶

	Type	Description
Field
id	integer	Airbnb's unique identifier for the listing
listing_url	text	NaN
scrape_id	bigint	Inside Airbnb "Scrape" this was part of
last_scraped	datetime	UTC. The date and time this listing was "scrap...
name	text	Name of the listing
description	text	Detailed description of the listing
neighborhood_overview	text	Host's description of the neighbourhood
picture_url	text	URL to the Airbnb hosted regular sized image f...
host_id	integer	Airbnb's unique identifier for the host/user
host_url	text	The Airbnb page for the host
host_name	text	Name of the host. Usually just the first name(s).
host_since	date	The date the host/user was created. For hosts ...
host_location	text	The host's self reported location
host_about	text	Description about the host
host_response_time	NaN	NaN
host_response_rate	NaN	NaN
host_acceptance_rate	NaN	That rate at which a host accepts booking requ...
host_is_superhost	boolean [t=true; f=false]	NaN
host_thumbnail_url	text	NaN
host_picture_url	text	NaN
host_neighbourhood	text	NaN
host_listings_count	text	The number of listings the host has (per Airbn...
host_total_listings_count	text	The number of listings the host has (per Airbn...
host_verifications	NaN	NaN
host_has_profile_pic	boolean [t=true; f=false]	NaN
host_identity_verified	boolean [t=true; f=false]	NaN
neighbourhood	text	NaN
neighbourhood_cleansed	text	The neighbourhood as geocoded using the latitu...
neighbourhood_group_cleansed	text	The neighbourhood group as geocoded using the ...
latitude	numeric	Uses the World Geodetic System (WGS84) project...
longitude	numeric	Uses the World Geodetic System (WGS84) project...
property_type	text	Self selected property type. Hotels and Bed an...
room_type	text	[Entire home/apt\|Private room\|Shared room\|Hote...
accommodates	integer	The maximum capacity of the listing
bathrooms	numeric	The number of bathrooms in the listing
bathrooms_text	string	The number of bathrooms in the listing. \nOn t...
bedrooms	integer	The number of bedrooms
beds	integer	The number of bed(s)
amenities	json	NaN
price	currency	daily price in local currency
minimum_nights	integer	minimum number of night stay for the listing (...
maximum_nights	integer	maximum number of night stay for the listing (...
minimum_minimum_nights	integer	the smallest minimum_night value from the cale...
maximum_minimum_nights	integer	the largest minimum_night value from the calen...
minimum_maximum_nights	integer	the smallest maximum_night value from the cale...
maximum_maximum_nights	integer	the largest maximum_night value from the calen...
minimum_nights_avg_ntm	numeric	the average minimum_night value from the calen...
maximum_nights_avg_ntm	numeric	the average maximum_night value from the calen...
calendar_updated	date	NaN
has_availability	boolean	[t=true; f=false]
availability_30	integer	avaliability_x. The availability of the listin...
availability_60	integer	avaliability_x. The availability of the listin...
availability_90	integer	avaliability_x. The availability of the listin...
availability_365	integer	avaliability_x. The availability of the listin...
calendar_last_scraped	date	NaN
number_of_reviews	integer	The number of reviews the listing has
number_of_reviews_ltm	integer	The number of reviews the listing has (in the ...
number_of_reviews_l30d	integer	The number of reviews the listing has (in the ...
first_review	date	The date of the first/oldest review
last_review	date	The date of the last/newest review
review_scores_rating	NaN	NaN
review_scores_accuracy	NaN	NaN
review_scores_cleanliness	NaN	NaN
review_scores_checkin	NaN	NaN
review_scores_communication	NaN	NaN
review_scores_location	NaN	NaN
review_scores_value	NaN	NaN
license	text	The licence/permit/registration number
instant_bookable	boolean	[t=true; f=false]. Whether the guest can autom...
calculated_host_listings_count	integer	The number of listings the host has in the cur...
calculated_host_listings_count_entire_homes	integer	The number of Entire home/apt listings the hos...
calculated_host_listings_count_private_rooms	integer	The number of Private room listings the host h...
calculated_host_listings_count_shared_rooms	integer	The number of Shared room listings the host ha...
reviews_per_month	numeric	The number of reviews the listing has over the...

	host_id	latitude	longitude	price	minimum_nights	number_of_reviews	reviews_per_month	calculated_host_listings_count	availability_365	number_of_reviews_ltm	...	beds	availability_60	number_of_reviews_l30d	availability_90	review_scores_accuracy	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value
count	3.809500e+04	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	...	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000	38095.000000
mean	9.369813e+07	48.863862	2.345970	122.423783	90.069720	28.052815	0.809993	9.625856	100.677175	4.717181	...	1.713112	13.537735	0.594277	23.171256	4.760658	4.596519	4.802792	4.813728	4.805951	4.618057
std	1.161688e+08	0.018109	0.032499	181.146475	160.968654	51.055294	1.197874	33.324839	132.730567	14.673778	...	1.409655	19.494251	1.559542	31.384957	0.421444	0.531285	0.397959	0.400307	0.348554	0.464299
min	2.626000e+03	48.812580	2.229290	0.000000	1.000000	1.000000	0.010000	1.000000	0.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	1.246941e+07	48.850790	2.325840	62.000000	2.000000	4.000000	0.130000	1.000000	0.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	4.700000	4.460000	4.760000	4.790000	4.750000	4.500000
50%	3.730673e+07	48.865110	2.348850	90.000000	3.000000	10.000000	0.380000	1.000000	7.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	4.880000	4.750000	4.930000	4.950000	4.920000	4.720000
75%	1.362256e+08	48.878235	2.370150	137.000000	30.000000	30.000000	1.000000	2.000000	203.000000	5.000000	...	2.000000	26.000000	0.000000	49.000000	5.000000	4.960000	5.000000	5.000000	5.000000	4.900000
max	4.346335e+08	48.904860	2.472030	9280.000000	9999.000000	1809.000000	50.860000	252.000000	365.000000	1705.000000	...	90.000000	60.000000	62.000000	90.000000	5.000000	5.000000	5.000000	5.000000	5.000000	5.000000

	neighbourhood	latitude	longitude	room_type	minimum_nights	number_of_reviews	reviews_per_month	calculated_host_listings_count	availability_365	number_of_reviews_ltm	...	beds	availability_60	number_of_reviews_l30d	availability_90	review_scores_accuracy	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value
count	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	...	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000	30016.000000
mean	9.199627	48.863914	2.346244	0.299973	90.901752	28.148354	0.813860	9.365505	99.526819	4.754864	...	1.678105	13.341784	0.595416	22.860641	4.760240	4.595398	4.802977	4.814794	4.804639	4.618969
std	6.010823	0.018180	0.032393	0.712868	163.525736	51.476934	1.213939	32.768867	132.153008	15.176982	...	1.268670	19.363373	1.573633	31.211762	0.421649	0.529750	0.395841	0.397186	0.347960	0.461740
min	0.000000	48.812580	2.229290	0.000000	1.000000	1.000000	0.010000	1.000000	0.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
25%	3.000000	48.850780	2.326290	0.000000	2.000000	4.000000	0.130000	1.000000	0.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	4.700000	4.460000	4.760000	4.790000	4.750000	4.500000
50%	9.000000	48.865217	2.349045	0.000000	3.000000	10.000000	0.380000	1.000000	6.000000	0.000000	...	1.000000	0.000000	0.000000	0.000000	4.880000	4.750000	4.930000	4.950000	4.910000	4.720000
75%	15.000000	48.878450	2.370350	0.000000	30.000000	30.000000	1.000000	2.000000	199.000000	5.000000	...	2.000000	26.000000	0.000000	48.000000	5.000000	4.960000	5.000000	5.000000	5.000000	4.890000
max	19.000000	48.904860	2.472030	3.000000	9999.000000	1809.000000	50.860000	252.000000	365.000000	1705.000000	...	83.000000	60.000000	62.000000	90.000000	5.000000	5.000000	5.000000	5.000000	5.000000	5.000000

	R2	MAE	MAPE
model
HistGradientBoosting	0.689298	27.107083	0.270250
LGBM	0.688734	27.068275	0.268787
xgboost	0.671849	27.900860	0.276963
RandomForest	0.657067	28.851235	0.294581
GradientBoosting	0.642329	29.056221	0.291377
Ridge	0.488570	35.495701	0.362573
LinearRegression	0.488565	35.506839	0.362786
Lasso	0.456026	36.403419	0.373664
LassoCV	0.431338	37.546634	0.394876
ElasticNetCV	0.341792	40.621188	0.449100
SVR	0.115485	44.554178	0.425445

	0	1	2	mean
LGBM	0.678723	0.666911	0.650251	0.665295
HistGradientBoosting	0.679939	0.667425	0.641856	0.663073
xgboost	0.659318	0.652688	0.610161	0.640722
GradientBoosting	0.641981	0.632034	0.617889	0.630635
RandomForest	0.646465	0.631082	0.577675	0.618407